[top] (x+11,960-x*1)/(x+11,960)^2

• Mjollnir · 12/08/07, 11:06 AM

Originally Posted by Satrina

Miss -> Dodge -> Parry -> Block -> Critical Strike -> Crushing Blow -> Hit

So 490def turns crits into hits, or crits into crushes? What would normally be a crit, negated by defense, becomes a ... ?

Satrina · 12/08/07, 11:26 AM

The chance gets converted to the lowest priority item on the table, so it becomes hit.

• frmorrison · 12/08/07, 2:55 PM

One error I saw:
Holy Paladin with improved devotion aura and set bonus

The Tier 5 set bonus changed to not give a bonus to Devo Aura.

One comment:
It seems bosses have different parry rates, similar to how bosses have different armor amounts.
Most bosses appear via WWS to have 11.2% parry, but a few appear to have 13%.

◊ Quigon · 12/08/07, 5:34 PM

I'm not going to change the parry comments until someone can provide concrete numbers. Parry % is not well known at the moment. I make that clear enough already.

Tymir · 12/08/07, 5:52 PM

Originally Posted by Quigon

I'm not going to change the parry comments until someone can provide concrete numbers. Parry % is not well known at the moment. I make that clear enough already.

I'm not sure if you have already seen this or not, but this post, while not a definitive answer to the boss parry rate, may aid in coming to a conclusion.

Sepulture · 12/08/07, 6:23 PM

Originally Posted by Edgewalker

I don't understand your concern.
There are really 0 situations you should have problems timing it correctly if you use vent or can type out simple sentences.

Sure sure. I used to argue your side of this debate until I got used to having the rage on the pull. On paper, it looks worthless (just like 3/3 imp TC when you could have an offtank do it) but it isn't.

The reason is, people make mistakes. Some more than others. In the case of my raids, well, we have significantly more somes rather than others.

◊ Quigon · 12/08/07, 7:03 PM

Originally Posted by Sepulture

Sure sure. I used to argue your side of this debate until I got used to having the rage on the pull. On paper, it looks worthless (just like 3/3 imp TC when you could have an offtank do it) but it isn't.

The reason is, people make mistakes. Some more than others. In the case of my raids, well, we have significantly more somes rather than others.

And yet making mistakes surely favors improved taunt more than improved bloodrage.
I've tried both myself - well over a year of having it both ways. I don't see how a tanking warrior can claim improved bloodrage is superior except when trash is not involved, or the boss does not have a mechanic or add or phase where taunt is useful. Furthermore, the most likely time to lose aggro is actually after having lost it and taunting the mob and the taunt debuff wears off.

Even before misdirection came about, who the lost aggro to bosses on a pull? That wouldn't be the tank's fault. Let's be realistic here. I've never had a problem having enough rage to shield slam on a boss pull.

The problem here is that people will not be objective about a spec that differs from their own. Improved bloodrage is great if nothing was tauntable, and we walked straight up to each of these untauntable bosses each night, and we were also losing aggro on the pull by about 1 melee swing in threat.

I can understand an ITC argument, even AM, but improved bloodrage just seems to be horribly devaluing improved taunt, which is one of our better talents.

All that being said, it is clearly stated in the OP that the choice is left to the user of improved bloodrage or improved taunt, and all of the opinion of the last page or so is left out (thankfully).

As to the parry post - I came up with a compilation of roughly 50,000 swings on Teron Gorefiend, but honestly the data is far too noisy to say anything for certain. Just look at how huge 1 more missed, or 1 less missed parry changes the shield slam data for maiden.

Sepulture · 12/09/07, 2:00 AM

On a different and useful note, (macros) I'd recommend adding a few lines where there is room. I don't see these in many peoples' macros.

/script UIErrorsFrame:Hide()
/script UIErrorsFrame:Clear(); UIErrorsFrame:Show()

They hide the three rows of irritating red error text that block your view and then reenable them at the end of your macro.

Here is a functioning example:

#show Shield Slam
/script UIErrorsFrame:Hide()
/startattack
/use [harm,nodead,equipped:Shields] 13
/stopcasting
/cast [equipped:Shields] Shield Slam
/script UIErrorsFrame:Clear(); UIErrorsFrame:Show()

You can also just macro

/script UIErrorsFrame:Hide()

if you want to leave them off for that session.

• Tauftamir · 12/09/07, 11:11 AM

Originally Posted by Sepulture

Sure sure. I used to argue your side of this debate until I got used to having the rage on the pull. On paper, it looks worthless (just like 3/3 imp TC when you could have an offtank do it) but it isn't.

The reason is, people make mistakes. Some more than others. In the case of my raids, well, we have significantly more somes rather than others.

So long as the key talents are noted in each build, then really it is up to each Warrior to decide on his or her own where to put the "filler" talents depending on how you are used to playing. There isn't a definitive answer as it essentially boils down to opinion on what is, or is not useful.

Note for those using a "Stance dance for Fear" macro mentioned earlier: I highly recommend unticking "Secure Ability Toggle" in the interface options, which is enabled by default. This adds a half second delay to triggering Berserker Rage when using a macro for this purpose.

I doubt the boss parry rate can be definitively pinned down due to parry caused by DPS from RSTS abilities or the boss casting spells. Does anyone know of a boss without such an ability that I could go and hit on for a few hours and get some data from?
Even if this is a possibility I don't think the results could be applied to all Boss mobs with 100% certainty.

Emrus · 12/09/07, 5:39 PM

About the Miss -> Dodge -> Parry -> Crushing Blow -> Critical Strike -> Hit question
From the old T2 bug it's certain that Parry precedes dodge, and have been shown multiple times.
About the Crush/crit order, back in the days pre tbc where 2.0 just came out with the lvl 70 ghosts at karazhan, there were quite a few people that tested the above out, and discovered that the ghosts never critted they did however crush about 80% of the time rest was miss/parry/dodge/block.

On to Armor
The formula for damage reduction against a lvl 73 boss is a given:

Now what we are interested is in finding out is the relative damage reduction percentage we gain pr armor.
Say if we go from 0 %-> 5% damage reduction we reduce the relative damage intake by 5%
If we go from 20%-> 25% ... by 6,5%
50%-> 55% by 10%
This relative reduction can be put on a formula:

However we don't get the same ΔDR% pr armor as our armor increases, and that's really what we are interested in finding. So lets figure out what the relative DR relative to our armor.

When we insert that in our formula we get:

And the Graph (percentage gain pr 100 armor)

As it's clearly visible the relative DR% is also subject to diminishing returns, it's not a whole lot in the 10000 - 35000 area but it's there. Note: The way DR% from armor is subject to diminishing returns is very alike the way stamina increase% is subject to diminishing returns since.

Since Itzlegend seems to be a solid source in this thread here's one by him.

That said, there is SOME diminishing returns to armor. It his however, extremly small and basically neglidgable.

And last Stamina Vs armor
One rating provies either 1,5 stamina or 10 armor
If a tank has 18000 hp and 15000 armor, each rating will provide with the above formulas:
RelativeHp increase pr 1,5 stam = 0.10%
RelativeArmor increase pr 10 armor = 0.04%

If a tank has 23000 hp and 18000 armor:
RelativeHp increase pr 1,5 stam = 0.075%
RelativeArmor increase pr 10 armor = 0.036%

While Armor only provides nearly half as much %increase as stamina it does however increase the damage reduction while at the same time increases survivability, while stamina only increases survivability.

note: English isn't my main language so my words might have been slightly misused. That doesn't change the result though.

Satrina · 12/09/07, 6:53 PM

Originally Posted by Emrus

About the Miss -> Dodge -> Parry -> Crushing Blow -> Critical Strike -> Hit question
From the old T2 bug it's certain that Parry precedes dodge, and have been shown multiple times.
About the Crush/crit order, back in the days pre tbc where 2.0 just came out with the lvl 70 ghosts at karazhan, there were quite a few people that tested the above out, and discovered that the ghosts never critted they did however crush about 80% of the time rest was miss/parry/dodge/block.

They certainly crit me when I was using them for testing at 60. The increased crush chance is as expected because we were 60 and the mobs were 70. 2% extra chance to crush per additional point of difference between attacker's weapon skill and defender's level maximum defense past 15. 80-ish percent chance to crush would be about right.

Edit: That said, if hellfire warders can have a screwy table, so can any mob. Could even be that the order is different for boss/not-boss for all we know. The order on the table is really only a point of interest since we blow away criticals via defense anyway.

◊ Quigon · 12/09/07, 7:52 PM

Just to expand on the calculus done above:
d/dx [x/(x+11960)]
= (x+11,960-x*(d/dx) [x[x+11,960])/(x+11,960)^2]
= (x+11,960-x*(d/dx) x[x])/(x+11,960)^2

[top] (x+11,960-x*1)/(x+11,960)^2

(11,960)/(x+11960)^2

Incidentally, including these infinitesimal points due to calculus actually makes the math for calculating relative DR worse by *about* a square factor. This means that it takes roughly 16 times as much armor to gain a point of DR at 75% armor as opposed to 0% armor - whereas you might expect only 4 times as much armor, since the gains in mitigation are about 4 fold.
At 50% it should be 4 times as much armor needed for a point of gain versus 0% armor, but each point of gain is only double that of 0%.

I'll add this once I can make everything look a bit prettier.

Very nice post Emrus.

• Zak · 12/10/07, 5:03 AM

I started using the Last Stand and Shield Wall bigwigs macros that throw up a timer when you use them, and really love them, many thanks. One question though, is there a way to stop it doing anything if the ability in question is on cooldown? Currently it still puts up the bigwigs bar even if it can't use the ability - my macroFu is lacking. For reference, here's the macro as it stands:

#showtooltip Last Stand
/cast Last Stand
/bwcb 20 Last Stand

zork · 12/10/07, 5:37 AM

Originally Posted by Zak

#showtooltip Last Stand
/cast Last Stand
/bwcb 20 Last Stand

I found the same function for deadlybossmods:
Deadly Boss Mods

Originally Posted by Zak

#showtooltip Last Stand
/cast Last Stand
/dbm broadcast timer 00:20 Last Stand

Should do it.

◊ Fellwraith · 12/10/07, 11:12 AM

Originally Posted by Quigon

As to the parry post - I came up with a compilation of roughly 50,000 swings on Teron Gorefiend, but honestly the data is far too noisy to say anything for certain. Just look at how huge 1 more missed, or 1 less missed parry changes the shield slam data for maiden.

Over the past 3 weeks I've seen the attached results from Hyjal and BT. I've missed a good chunk of the easier farm bosses in BT because I was working late and/or forgot to start logging until later that night. I'm not including Gurtogg because there are many occassions where I will not be facing him. I'm not including data from older parses because it's either pre-2.3 or I didn't have the same expertise rating I have now.

I have -8% to enemy dodge and parry in every parse, however, my hit will vary based on the boss (I started using Pepe's for trash and squishier bosses like Rage Winterchill). If the parry rate were 11.2% I'd see a much lower net parry in a sample like this. This is especially true once you factor in channeled spells, hateful strikes, and instant-cast & turn spells (which would prevent them from parrying periodically).

Hudini · 12/10/07, 1:08 PM

Please add to the guide that 3/3 Tactical Mastery causes Mortal Strike and Bloodthrist to do 63% of their damage in additional threat.

I realize this thread is primarily targeted toward deep protection specs, but I have been trying out a 0/46/15 hybrid spec and have found my threat is still very competitive, especially if I am partied in the melee DPS group with a feral druid and enhance shammy.

My guild (currently on Vashj) is running many raids very tank heavy, which ends up hurting our DPS, so I'm giving this spec a whirl to try to give our raids a little more flexibility. So far I'm finding that this hybrid build works very well for both damage and threat if the warrior has the gear and group composition to support both roles. 10 man raids seem specifically designed to have a tank/offspec-tank line-ups, where the offspec switches gear depending on the boss or trash.

I haven't tabulated hard numbers, but anecdotally:

If I gear with emphasis on damage and minimum survivability (490 defense, high stam, high AP) and have group synergy, I'm getting threat numbers higher than I ever got with deep prot with the same group composition. If I gear for maximum survivability in the same group composition, I'm pulling roughly 800-900 threat per second, which is probably 200 TPS lower than what I was capable of as deep prot with no group synergy.

This spec feels acceptable for off-tank duties with respectable damage when needed.

tangent · 12/10/07, 1:58 PM

Be careful with the Shieldwall bigwigs bar. The bar will go up even if you hit shield wall during a global cooldown.
If you end up getting a before the GCD comes up and decide that shield wall isn't needed you still have a bar up telling your healers that you're shield walled.

• Xaviera · 12/10/07, 4:16 PM

Originally Posted by tangent

Be careful with the Shieldwall bigwigs bar. The bar will go up even if you hit shield wall during a global cooldown.
If you end up getting a before the GCD comes up and decide that shield wall isn't needed you still have a bar up telling your healers that you're shield walled.

Yeah. If you still want a notifier for your raid/yourself on your Shield Wall / Last Stand usage though, Tankalyze, an Ace mod, works great. I've used it for a while, and you can configure it to show a variety of different raid broadcasted messages, and when they're fading. I only have mine set to announce the initial usage of the timers though, and also to broadcast taunt/aoe taunt/mocking blow resists. (As well as announcing the use of AOE taunt). It's a handy, quick mod that I highly recommend.

◊ Quigon · 12/10/07, 5:26 PM

Originally Posted by Fellwraith

Over the past 3 weeks I've seen the attached results from Hyjal and BT. I've missed a good chunk of the easier farm bosses in BT because I was working late and/or forgot to start logging until later that night. I'm not including Gurtogg because there are many occassions where I will not be facing him. I'm not including data from older parses because it's either pre-2.3 or I didn't have the same expertise rating I have now.

I have -8% to enemy dodge and parry in every parse, however, my hit will vary based on the boss (I started using Pepe's for trash and squishier bosses like Rage Winterchill). If the parry rate were 11.2% I'd see a much lower net parry in a sample like this. This is especially true once you factor in channeled spells, hateful strikes, and instant-cast & turn spells (which would prevent them from parrying periodically).

Seems like around 13-14%
This matches what I see as well... basically that it is nearly 15% parry rate, and sometimes before expertise implementation, having over 20% parry rate on a boss was not uncommon.
It is definitely likely that some bosses have higher parry rates than others, but I think simply that boss parry rate is somewhat high. I don't know what the explanation could be there. Maybe its just a flat 15% - but that seems TOO high.
I'm glad to see dodge disappear for you.

This is a really nice addition.

Edit:
In fact, I think it is almost safe to assume 15% is the parry rate - and the actual "seen" of less than 15% might be due to the issues you just described. I'll add something to this effect when I get a chance later this week to update the guide.

PrayForDeath · 12/10/07, 8:46 PM

Excellent post Quigon; thank you very much for taking the time to create this comprehensive guide. I have a few comments:

Could you explain the relation between Shield Block Value and Shield Slam?

Could you explain why fast weapons are preferable for tanking?

And finally, I suggest that you mention how 1HWS affects shield slam as well.

I know that most experienced tanks already know the answers to those questions, but I'm sure it would be useful to include this info in the guide for people who are new to the class/raiding.

Dralmoo · 12/12/07, 1:55 AM

[Onslaught Handguards] vs [Royal Gauntlets of Silvermoon] - Are the T6 gloves really better? I've passed on the T6 gloves a few times now - by wearing them I'd lose 22 Stam and 5 Dodge for 8 def and 60 armor, discounting the (mostly useless) Strength and shield block rating. Doesn't seem worth it...

Although, the set bonus is a non-factor for me as the T6 Helm is also not so great and it will be quite a while before i see the shoulders..

Ihmes · 12/12/07, 2:40 AM

Originally Posted by Dralmoo

[Onslaught Handguards] vs